derek-thomas's picture
derek-thomas HF staff
Update README.md
ff905a3
metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
  - generated_from_trainer
datasets:
  - facebook/voxpopuli
  - voxpopuli/hr
model-index:
  - name: speecht5_finetuned_voxpopuli_hr
    results: []
pipeline_tag: text-to-speech
language:
  - hr

speecht5_finetuned_voxpopuli_it

This model is a fine-tuned version of microsoft/speecht5_tts on the voxpopuli/it dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4497

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 4
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 4000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.4773 32.52 1000 0.4502
0.4545 65.04 2000 0.4462
0.4488 97.56 3000 0.4502
0.4517 130.08 4000 0.4497

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0